Author: "Facebook AI Research [Paris] (FAIR)" / Topic: 0202 electrical engineering, electronic engineering, information engineering - Searchworks@Jio Institute Digital Library Search Results

Your search keyword '"Facebook AI Research [Paris] (FAIR)"' showing total 41 results

Start Over Author "Facebook AI Research [Paris] (FAIR)" Topic 0202 electrical engineering, electronic engineering, information engineering

41 results on '"Facebook AI Research [Paris] (FAIR)"'

1. Black-Box Optimization Revisited: Improving Algorithm Selection Wizards Through Massive Benchmarking

Author: Laurent Meunier, Paco Wong, Jeremy Rapin, Antoine Moreau, Olivier Teytaud, Carola Doerr, Baptiste Roziere, Herilalaina Rakotoarison, Facebook AI Research [Paris] (FAIR), Facebook, Laboratoire Interdisciplinaire des Sciences du Numérique (LISN), CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), TAckling the Underspecified (TAU), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Interdisciplinaire des Sciences du Numérique (LISN), CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), The Chinese University of Hong Kong [Hong Kong], Université Clermont Auvergne (UCA), SIGMA Clermont (SIGMA Clermont), Institut Pascal (IP), Centre National de la Recherche Scientifique (CNRS)-Université Clermont Auvergne (UCA)-Institut national polytechnique Clermont Auvergne (INP Clermont Auvergne), Université Clermont Auvergne (UCA)-Université Clermont Auvergne (UCA), Recherche Opérationnelle (RO), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Laboratoire d'informatique de l'École polytechnique [Palaiseau] (LIX), École polytechnique (X)-Centre National de la Recherche Scientifique (CNRS), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), and Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, Optimization problem, Computer science, 02 engineering and technology, Overfitting, Machine learning, computer.software_genre, Machine Learning (cs.LG), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Theoretical Computer Science, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], Black box, 0202 electrical engineering, electronic engineering, information engineering, business.industry, Black-Box Optimization, Suite, Benchmarking, Wizard, Range (mathematics), Computational Theory and Mathematics, Benchmark (computing), 020201 artificial intelligence & image processing, Artificial intelligence, business, computer, Software
Abstract: International audience; Existing studies in black-box optimization for machine learning suffer from low generalizability, caused by a typically selective choice of problem instances used for training and testing different optimization algorithms. Among other issues, this practice promotes overfitting and poor-performing user guidelines. To address this shortcoming, we propose in this work a benchmark suite, OptimSuite, which covers a broad range of black-box optimization problems, ranging from academic benchmarks to real-world applications, from discrete over numerical to mixed-integer problems, from small to very large-scale problems, from noisy over dynamic to static problems, etc. We demonstrate the advantages of such a broad collection by deriving from it Automated Black Box Optimizer (ABBO), a general-purpose algorithm selection wizard. Using three different types of algorithm selection techniques, ABBO achieves competitive performance on all benchmark suites. It significantly outperforms previous state of the art on some of them, including YABBOB and LSGO. ABBO relies on many high-quality base components. Its excellent performance is obtained without any task-specific parametrization. The OptimSuite benchmark collection, the ABBO wizard and its base solvers have all been merged into the open-source Nevergrad platform, where they are available for reproducible research.
Published: 2022
Full Text: View/download PDF

2. Asymptotic convergence rates for averaging strategies

Author: Iskander Legheraba, Yann Chevaleyre, Laurent Meunier, Olivier Teytaud, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Facebook AI Research [Paris] (FAIR), and Facebook
Subjects: FOS: Computer and information sciences, Design of experiments, Computer Science - Neural and Evolutionary Computing, 0102 computer and information sciences, 02 engineering and technology, Function (mathematics), 01 natural sciences, Evolutionary computation, Random search, Quadratic equation, 010201 computation theory & mathematics, Optimization and Control (math.OC), Black box, Convergence (routing), 0202 electrical engineering, electronic engineering, information engineering, FOS: Mathematics, Applied mathematics, 020201 artificial intelligence & image processing, [INFO]Computer Science [cs], Neural and Evolutionary Computing (cs.NE), [MATH]Mathematics [math], Mathematics - Optimization and Control, Arithmetic mean, Mathematics
Abstract: Parallel black box optimization consists in estimating the optimum of a function using $\lambda$ parallel evaluations of $f$. Averaging the $\mu$ best individuals among the $\lambda$ evaluations is known to provide better estimates of the optimum of a function than just picking up the best. In continuous domains, this averaging is typically just based on (possibly weighted) arithmetic means. Previous theoretical results were based on quadratic objective functions. In this paper, we extend the results to a wide class of functions, containing three times continuously differentiable functions with unique optimum. We prove formal rate of convergences and show they are indeed better than pure random search asymptotically in $\lambda$. We validate our theoretical findings with experiments on some standard black box functions.
Published: 2021

3. The Zero Resource Speech Challenge 2021: Spoken Language Modelling

Author: Ewan Dunbar, Emmanuel Dupoux, Mathieu Bernard, Patricia Rozé, Maureen de Seyssel, Morgane Riviere, Tu Anh Nguyen, Nicolas Hamilakis, Eugene Kharitonov, University of Toronto, Laboratoire de sciences cognitives et psycholinguistique (LSCP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), Apprentissage machine et développement cognitif (CoML), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Département d'Etudes Cognitives - ENS Paris (DEC), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Facebook AI Research [Paris] (FAIR), Facebook, ANR-19-P3IA-0001,PRAIRIE,PaRis Artificial Intelligence Research InstitutE(2019), ANR-17-EURE-0017,FrontCog,Frontières en cognition(2017), ANR-10-IDEX-0001,PSL,Paris Sciences et Lettres(2010), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de sciences cognitives et psycholinguistique (LSCP), École normale supérieure - Paris (ENS-PSL), and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL)
Subjects: FOS: Computer and information sciences, Computer Science - Artificial Intelligence, Computer science, 02 engineering and technology, computer.software_genre, Zero-resource, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Standard language, 030507 speech-language pathology & audiology, 03 medical and health sciences, Similarity (psychology), 0202 electrical engineering, electronic engineering, information engineering, Cognitive benchmarks, Computer Science - Computation and Language, business.industry, 020206 networking & telecommunications, computer.file_format, Pipeline (software), Unsupervised speech, Zero (linguistics), Artificial Intelligence (cs.AI), Language modelling, Lowresource, Language model, Artificial intelligence, ABX test, 0305 other medical science, business, Computation and Language (cs.CL), computer, Encoder, Natural language processing, Spoken language
Abstract: We present the Zero Resource Speech Challenge 2021, which asks participants to learn a language model directly from audio, without any text or labels. The challenge is based on the Libri-light dataset, which provides up to 60k hours of audio from English audio books without any associated text. We provide a pipeline baseline system consisting on an encoder based on contrastive predictive coding (CPC), a quantizer ($k$-means) and a standard language model (BERT or LSTM). The metrics evaluate the learned representations at the acoustic (ABX discrimination), lexical (spot-the-word), syntactic (acceptability judgment) and semantic levels (similarity judgment). We present an overview of the eight submitted systems from four groups and discuss the main results., Submitted to Interspeech 2021. arXiv admin note: text overlap with arXiv:2011.11588
Published: 2021
Full Text: View/download PDF

4. Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering

Author: Edouard Grave, Gautier Izacard, Facebook AI Research [Paris] (FAIR), Facebook, Statistical Machine Learning and Parsimony (SIERRA), Département d'informatique - ENS Paris (DI-ENS), Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Département d'informatique de l'École normale supérieure (DI-ENS), École normale supérieure - Paris (ENS Paris), and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS Paris)
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, Computer Science - Computation and Language, Information retrieval, Computer science, 05 social sciences, Aggregate (data warehouse), 02 engineering and technology, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], Machine Learning (cs.LG), 020204 information systems, 0202 electrical engineering, electronic engineering, information engineering, Open domain, Question answering, Natural (music), 0509 other social sciences, 050904 information & library sciences, Computation and Language (cs.CL), Generative grammar
Abstract: International audience; Generative models for open domain question answering have proven to be competitive, without resorting to external knowledge. While promising, this approach requires to use models with billions of parameters, which are expensive to train and query. In this paper, we investigate how much these models can benefit from retrieving text passages, potentially containing evidence. We obtain state-of-the-art results on the Natural Questions and Triv-iaQA open benchmarks. Interestingly, we observe that the performance of this method significantly improves when increasing the number of retrieved passages. This is evidence that sequence-to-sequence models offers a flexible framework to efficiently aggregate and combine evidence from multiple passages.
Published: 2021
Full Text: View/download PDF

5. Nevergrad: black-box optimization platform

Author: Fabien Teytaud, Antoine Moreau, Pauline Bennet, Olivier Teytaud, Carola Doerr, Jeremy Rapin, Institut Pascal (IP), Centre National de la Recherche Scientifique (CNRS)-Université Clermont Auvergne (UCA)-Institut national polytechnique Clermont Auvergne (INP Clermont Auvergne), Université Clermont Auvergne (UCA)-Université Clermont Auvergne (UCA), Facebook AI Research [Paris] (FAIR), Facebook, Université Paris-Saclay, Laboratoire d'Informatique Signal et Image de la Côte d'Opale (LISIC), Université du Littoral Côte d'Opale (ULCO), and Université de Lille
Subjects: Black box (phreaking), Computer science, 0102 computer and information sciences, 02 engineering and technology, General Medicine, computer.software_genre, 01 natural sciences, Open source, 010201 computation theory & mathematics, User group, 0202 electrical engineering, electronic engineering, information engineering, Operating system, [SPI.OPTI]Engineering Sciences [physics]/Optics / Photonic, Join (sigma algebra), 020201 artificial intelligence & image processing, computer
Abstract: Nevergrad is an open source platform for black-box optimization. Join the user group! And if you like Nevergrad, please support us by adding a star on GitHub (https://github.com/facebookresearch/nevergrad).
Published: 2021
Full Text: View/download PDF

6. EvolGAN: Evolutionary Generative Adversarial Networks

Author: Fabien Teytaud, Jeremy Rapin, Hanhe Lin, Baptiste Roziere, Vlad Hosu, Olivier Teytaud, Mariia Zameshina, Facebook AI Research [Paris] (FAIR), Facebook, Laboratoire d'Informatique Signal et Image de la Côte d'Opale (LISIC), Université du Littoral Côte d'Opale (ULCO), Limnological Institute, University of Konstanz, Konstanz, Germany, Université Grenoble Alpes - UFR Informatique et Mathématiques Appliquées (UGA UFR IMAG), Université Grenoble Alpes (UGA), and Teytaud, Fabien
Subjects: [INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], FOS: Computer and information sciences, Computer Science - Machine Learning, Computer science, Computer Vision and Pattern Recognition (cs.CV), media_common.quotation_subject, Computer Science - Computer Vision and Pattern Recognition, 02 engineering and technology, 010501 environmental sciences, Space (commercial competition), 01 natural sciences, Machine Learning (cs.LG), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Image (mathematics), Adversarial system, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 0202 electrical engineering, electronic engineering, information engineering, Quality (business), ComputingMilieux_MISCELLANEOUS, 0105 earth and related environmental sciences, media_common, business.industry, Estimator, [INFO.INFO-LG] Computer Science [cs]/Machine Learning [cs.LG], 020201 artificial intelligence & image processing, Artificial intelligence, business, Generative grammar, Generator (mathematics)
Abstract: We propose to use a quality estimator and evolutionary methods to search the latent space of generative adversarial networks trained on small, difficult datasets, or both. The new method leads to the generation of significantly higher quality images while preserving the original generator's diversity. Human raters preferred an image from the new version with frequency 83.7pc for Cats, 74pc for FashionGen, 70.4pc for Horses, and 69.2pc for Artworks, and minor improvements for the already excellent GANs for faces. This approach applies to any quality scorer and GAN generator., accepted ACCV oral
Published: 2021
Full Text: View/download PDF

7. Towards unsupervised learning of speech features in the wild

Author: Morgane Riviere, Emmanuel Dupoux, Facebook AI Research [Paris] (FAIR), Facebook, Laboratoire de sciences cognitives et psycholinguistique (LSCP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), Apprentissage machine et développement cognitif (CoML), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Département d'Etudes Cognitives - ENS Paris (DEC), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), CIFAR program in Learning in Machines & Brains CIFAR LMB program, ANR-19-P3IA-0001,PRAIRIE,PaRis Artificial Intelligence Research InstitutE(2019), ANR-17-EURE-0017,FrontCog,Frontières en cognition(2017), ANR-10-IDEX-0001,PSL,Paris Sciences et Lettres(2010), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), Inria de Paris, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de sciences cognitives et psycholinguistique (LSCP)
Subjects: Voice activity detection, Perplexity, Computer science, Speaker adaptation, Speech recognition, Contrastive predictive coding, 02 engineering and technology, computer.file_format, 010501 environmental sciences, Unsupervised representation learning, 01 natural sciences, Task (project management), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Data filtering, 0202 electrical engineering, electronic engineering, information engineering, Task analysis, Unsupervised learning, 020201 artificial intelligence & image processing, [INFO]Computer Science [cs], ABX test, Representation (mathematics), Set (psychology), computer, 0105 earth and related environmental sciences
Abstract: International audience; Recent work on unsupervised contrastive learning of speech representation has shown promising results, but so far has mostly been applied to clean, curated speech datasets. Can it also be used with unprepared audio data "in the wild"? Here, we explore three potential problems in this setting: (i) presence of non-speech data, (ii) noisy or low quality speech data, and (iii) imbalance in speaker distribution. We show that on the Libri-light train set, which is itself a relatively clean speech-only dataset, these problems combined can already have a performance cost of up to 30% relative for the ABX score. We show that the first two problems can be alleviated by data filtering, with voice activity detection selecting speech segments, while perplexity of a model trained with clean data helping to discard entire files. We show that the third problem can be alleviated by learning a speaker embedding in the predictive branch of the model. We show that these techniques build more robust speech features that can be transferred to an ASR task in the low resource setting.
Published: 2020

8. Multilingual AMR-to-Text Generation

Author: Angela Fan, Claire Gardent, Facebook AI Research [Paris] (FAIR), Facebook, Natural Language Processing : representations, inference and semantics (SYNALP), Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), and Fan, Angela
Subjects: FOS: Computer and information sciences, Morphology (linguistics), Computer Science - Computation and Language, Computer science, business.industry, 02 engineering and technology, 010501 environmental sciences, computer.software_genre, 01 natural sciences, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], Bridging (programming), [INFO.INFO-CL] Computer Science [cs]/Computation and Language [cs.CL], 0202 electrical engineering, electronic engineering, information engineering, Text generation, ComputingMethodologies_DOCUMENTANDTEXTPROCESSING, Leverage (statistics), 020201 artificial intelligence & image processing, Artificial intelligence, business, computer, Computation and Language (cs.CL), Natural language, Natural language processing, 0105 earth and related environmental sciences, Word order
Abstract: International audience; Generating text from structured data is challenging because it requires bridging the gap between (i) structure and natural language (NL) and (ii) semantically underspecified input and fully specified NL output. Multilingual generation brings in an additional challenge: that of generating into languages with varied word order and morphological properties. In this work, we focus on Abstract Meaning Representations (AMRs) as structured input, where previous research has overwhelmingly focused on generating only into English. We leverage advances in cross-lingual embeddings, pretraining, and multilingual models to create multilingual AMR-to-text models that generate in twenty one different languages. For eighteen languages, based on automatic metrics, our multilingual models surpass baselines that generate into a single language. We analyse the ability of our multilingual models to accurately capture morphology and word order using human evaluation, and find that native speakers judge our generations to be fluent.
Published: 2020

9. Efficient Wait-k Models for Simultaneous Machine Translation

Author: Maha Elbayad, Laurent Besacier, Jakob Verbeek, Laboratoire d'Informatique de Grenoble (LIG), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), Apprentissage de modèles à partir de données massives (Thoth), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK), Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Institut Universitaire de France (IUF), Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.), Groupe d’Étude en Traduction Automatique/Traitement Automatisé des Langues et de la Parole (GETALP), Facebook AI Research [Paris] (FAIR), Facebook, and ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019)
Subjects: FOS: Computer and information sciences, Sound (cs.SD), 0209 industrial biotechnology, Computer Science - Computation and Language, Machine translation, Computer science, 02 engineering and technology, computer.software_genre, Computer Science - Sound, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], 020901 industrial engineering & automation, Computer engineering, Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Computation and Language (cs.CL), computer, Encoder, Decoding methods, Electrical Engineering and Systems Science - Audio and Speech Processing, Transformer (machine learning model), Spoken language
Abstract: Simultaneous machine translation consists in starting output generation before the entire input sequence is available. Wait-k decoders offer a simple but efficient approach for this problem. They first read k source tokens, after which they alternate between producing a target token and reading another source token. We investigate the behavior of wait-k decoding in low resource settings for spoken corpora using IWSLT datasets. We improve training of these models using unidirectional encoders, and training across multiple values of k. Experiments with Transformer and 2D-convolutional architectures show that our wait-k models generalize well across a wide range of latency levels. We also show that the 2D-convolution architecture is competitive with Transformers for simultaneous translation of spoken language., Accepted at INTERSPEECH 2020
Published: 2020
Full Text: View/download PDF

10. On Averaging the Best Samples in Evolutionary Computation

Author: Olivier Teytaud, Laurent Meunier, Clément W. Royer, Yann Chevaleyre, Jeremy Rapin, Facebook AI Research [Paris] (FAIR), Facebook, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Machine Intelligence and Learning Systems (MILES), and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL
Subjects: Selection (relational algebra), Order (ring theory), Regret, 0102 computer and information sciences, 02 engineering and technology, Lambda, 01 natural sciences, Evolutionary computation, Combinatorics, [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], 010201 computation theory & mathematics, Simple (abstract algebra), 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Sphere function, ComputingMilieux_MISCELLANEOUS, Mathematics
Abstract: Choosing the right selection rate is a long standing issue in evolutionary computation. In the continuous unconstrained case, we prove mathematically that a single parent $\mu =1$ leads to a sub-optimal simple regret in the case of the sphere function. We provide a theoretically-based selection rate $\mu /\lambda $ that leads to better progress rates. With our choice of selection rate, we get a provable regret of order $O(\lambda ^{-1})$ which has to be compared with $O(\lambda ^{-2/d})$ in the case where $\mu =1$. We complete our study with experiments to confirm our theoretical claims.
Published: 2020
Full Text: View/download PDF

11. Discrete Point Flow Networks for Efficient Point Cloud Generation

Author: Roman Klokov, Jakob Verbeek, Edmond Boyer, Apprentissage de modèles à partir de données massives (Thoth), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK), Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), Capture and Analysis of Shapes in Motion (MORPHEO), Facebook AI Research [Paris] (FAIR), Facebook, and ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019)
Subjects: FOS: Computer and information sciences, Speedup, Computer science, Computer Vision and Pattern Recognition (cs.CV), Computer Science - Computer Vision and Pattern Recognition, Point cloud, Inference, 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, 3D shape modeling, normalizing flows, 0202 electrical engineering, electronic engineering, information engineering, Point (geometry), [INFO]Computer Science [cs], Latent variable model, Representation (mathematics), 0105 earth and related environmental sciences, generative modeling, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], point cloud generation, Flow (mathematics), 020201 artificial intelligence & image processing, Affine transformation, Algorithm, single view reconstruction
Abstract: Generative models have proven effective at modeling 3D shapes and their statistical variations. In this paper we investigate their application to point clouds, a 3D shape representation widely used in computer vision for which, however, only few generative models have yet been proposed. We introduce a latent variable model that builds on normalizing flows with affine coupling layers to generate 3D point clouds of an arbitrary size given a latent shape representation. To evaluate its benefits for shape modeling we apply this model for generation, autoencoding, and single-view shape reconstruction tasks. We improve over recent GAN-based models in terms of most metrics that assess generation and autoencoding. Compared to recent work based on continuous flows, our model offers a significant speedup in both training and inference times for similar or better performance. For single-view shape reconstruction we also obtain results on par with state-of-the-art voxel, point cloud, and mesh-based methods., In ECCV'20
Published: 2020
Full Text: View/download PDF

12. Versatile black-box optimization

Author: Mike Preuss, Baptiste Roziere, Antoine Moreau, Fabien Teytaud, Jeremy Rapin, Jialin Liu, Olivier Teytaud, Southern University of Science and Technology of China (SUSTech), Université Clermont Auvergne [2017-2020] (UCA [2017-2020]), Leiden Institute of Advanced Computer Science [Leiden] (LIACS), Universiteit Leiden [Leiden], Facebook AI Research [Paris] (FAIR), Facebook, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Laboratoire d'Informatique Signal et Image de la Côte d'Opale (LISIC), Université du Littoral Côte d'Opale (ULCO), Southern University of Science and Technology (SUSTech), and Universiteit Leiden
Subjects: FOS: Computer and information sciences, Computer science, Computer Science - Artificial Intelligence, Testbed, Evolutionary algorithm, Computer Science - Neural and Evolutionary Computing, 0102 computer and information sciences, 02 engineering and technology, [INFO.INFO-RO]Computer Science [cs]/Operations Research [cs.RO], 01 natural sciences, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Artificial Intelligence (cs.AI), 010201 computation theory & mathematics, Black box, Component (UML), 0202 electrical engineering, electronic engineering, information engineering, Combinatorial optimization, 020201 artificial intelligence & image processing, Neural and Evolutionary Computing (cs.NE), [MATH.MATH-OC]Mathematics [math]/Optimization and Control [math.OC], Algorithm, ComputingMilieux_MISCELLANEOUS
Abstract: Choosing automatically the right algorithm using problem descriptors is a classical component of combinatorial optimization. It is also a good tool for making evolutionary algorithms fast, robust and versatile. We present Shiwa, an algorithm good at both discrete and continuous, noisy and noise-free, sequential and parallel, black-box optimization. Our algorithm is experimentally compared to competitors on YABBOB, a BBOB comparable testbed, and on some variants of it, and then validated on several real world testbeds., Accepted at GECCO 2020
Published: 2020
Full Text: View/download PDF

13. Unsupervised pretraining transfers well across languages

Author: Pierre-Emmanuel Mazaré, Emmanuel Dupoux, Morgane Riviere, Armand Joulin, Facebook AI Research [Paris] (FAIR), Facebook, Apprentissage machine et développement cognitif (CoML), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de sciences cognitives et psycholinguistique (LSCP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Département d'Etudes Cognitives - ENS Paris (DEC), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), Laboratoire de sciences cognitives et psycholinguistique (LSCP), CIFAR program in Learning in Machines & Brains CIFAR LMB program, ANR-19-P3IA-0001,PRAIRIE,PaRis Artificial Intelligence Research InstitutE(2019), ANR-17-EURE-0017,FrontCog,Frontières en cognition(2017), ANR-10-IDEX-0001,PSL,Paris Sciences et Lettres(2010), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Sound (cs.SD), Computer Science - Machine Learning, Computer science, InformationSystems_INFORMATIONSTORAGEANDRETRIEVAL, 02 engineering and technology, 010501 environmental sciences, computer.software_genre, 01 natural sciences, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], Computer Science - Sound, Machine Learning (cs.LG), Transcription (linguistics), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, 0202 electrical engineering, electronic engineering, information engineering, 0105 earth and related environmental sciences, Computer Science - Computation and Language, business.industry, Low resources, 020206 networking & telecommunications, ComputingMethodologies_PATTERNRECOGNITION, Unsupervised pretraining, [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], Artificial intelligence, business, Computation and Language (cs.CL), computer, Natural language processing, Cross-lingual, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract: Cross-lingual and multi-lingual training of Automatic Speech Recognition (ASR) has been extensively investigated in the supervised setting. This assumes the existence of a parallel corpus of speech and orthographic transcriptions. Recently, contrastive predictive coding (CPC) algorithms have been proposed to pretrain ASR systems with unlabelled data. In this work, we investigate whether unsupervised pretraining transfers well across languages. We show that a slight modification of the CPC pretraining extracts features that transfer well to other languages, being on par or even outperforming supervised pretraining. This shows the potential of unsupervised methods for languages with few linguistic resources., 6 pages. Accepted at ICASSP 2020. However the 2 pages of supplementary materials will appear only in the arxiv version
Published: 2020
Full Text: View/download PDF

14. Online Versus Offline NMT Quality: An In-depth Analysis on English-German and German-English

Author: Emmanuelle Esperança-Rodier, Laurent Besacier, Jakob Verbeek, Francis Brunet-Manquat, Maha Elbayad, Michael Ustaszewski, Laboratoire d'Informatique de Grenoble (LIG), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), Apprentissage de modèles à partir de données massives (Thoth), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK), Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Groupe d’Étude en Traduction Automatique/Traitement Automatisé des Langues et de la Parole (GETALP), Universität Innsbruck [Innsbruck], Facebook AI Research [Paris] (FAIR), Facebook, Institut Universitaire de France (IUF), Ministère de l'Education nationale, de l’Enseignement supérieur et de la Recherche (M.E.N.E.S.R.), and ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019)
Subjects: FOS: Computer and information sciences, Computer Science - Computation and Language, Machine translation, Computer science, business.industry, 02 engineering and technology, 010501 environmental sciences, computer.software_genre, 01 natural sciences, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], language.human_language, German, 0202 electrical engineering, electronic engineering, information engineering, language, 020201 artificial intelligence & image processing, Artificial intelligence, business, Computation and Language (cs.CL), computer, Natural language processing, Decoding methods, 0105 earth and related environmental sciences, Transformer (machine learning model)
Abstract: We conduct in this work an evaluation study comparing offline and online neural machine translation architectures. Two sequence-to-sequence models: convolutional Pervasive Attention (Elbayad et al. 2018) and attention-based Transformer (Vaswani et al. 2017) are considered. We investigate, for both architectures, the impact of online decoding constraints on the translation quality through a carefully designed human evaluation on English-German and German-English language pairs, the latter being particularly sensitive to latency constraints. The evaluation results allow us to identify the strengths and shortcomings of each model when we shift to the online setup., Comment: Accepted at COLING 2020
Published: 2020
Full Text: View/download PDF

15. Variance Reduction for Better Sampling in Continuous Domains

Author: Laurent Meunier, Jeremy Rapin, Carola Doerr, Olivier Teytaud, Facebook AI Research [Paris] (FAIR), Facebook, Centre National de la Recherche Scientifique (CNRS), Recherche Opérationnelle (RO), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), and Sorbonne Université (SU)
Subjects: FOS: Computer and information sciences, 050101 languages & linguistics, Computer Science - Machine Learning, Population, Initialization, Sample (statistics), Machine Learning (stat.ML), 02 engineering and technology, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Machine Learning (cs.LG), Random search, Statistics - Machine Learning, Prior probability, 0202 electrical engineering, electronic engineering, information engineering, 0501 psychology and cognitive sciences, Neural and Evolutionary Computing (cs.NE), education, Mathematics, education.field_of_study, 05 social sciences, Sampling (statistics), Computer Science - Neural and Evolutionary Computing, Probability distribution, 020201 artificial intelligence & image processing, Variance reduction, Algorithm
Abstract: International audience; Design of experiments, random search, initialization of population-based methods, or sampling inside an epoch of an evolutionary algorithm use a sample drawn according to some probability distribution for approximating the location of an optimum. Recent papers have shown that the optimal search distribution, used for the sampling, might be more peaked around the center of the distribution than the prior distribution modelling our uncertainty about the location of the optimum. We confirm this statement, provide explicit values for this reshaping of the search distribution depending on the population size λ and the dimension d, and validate our results experimentally.
Published: 2020
Full Text: View/download PDF

16. Artificial Intelligence for Games

Author: Bruno Bouzy, Vincent Corruble, Tristan Cazenave, Olivier Teytaud, Laboratoire d'Informatique Paris Descartes (LIPADE (URP_2517)), Université de Paris (UP), Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Systèmes Multi-Agents (SMA), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Facebook AI Research [Paris] (FAIR), and Facebook
Subjects: 010201 computation theory & mathematics, Computer science, business.industry, Monte Carlo tree search, 0202 electrical engineering, electronic engineering, information engineering, ComputingMilieux_PERSONALCOMPUTING, 020201 artificial intelligence & image processing, [INFO]Computer Science [cs], 0102 computer and information sciences, 02 engineering and technology, Artificial intelligence, business, 01 natural sciences
Abstract: International audience; This chapter presents the classical alpha-beta algorithm and several variants, Monte Carlo Tree Search which is at the origin of recent progresses in many games, techniques used in video games and puzzles, and retrograde analysis which performs perfect play in endgames.
Published: 2020
Full Text: View/download PDF

17. Polygames: Improved Zero Learning

Author: Vasil Khalidov, Gabriel Synnaeve, Qucheng Gong, Fabien Teytaud, Hengyuan Hu, Vegard Mella, Olivier Teytaud, Xian-Dong Chiu, Maria Elsa, Shi-Yu Chen, Cheng-Ling Li, Baptiste Roziere, Xavier Martinet, Guan-Wei Chen, Yi-Jun Ye, Yen-Chi Chen, Tristan Cazenave, Hsin-I Lin, Jeremy Rapin, Shi-Cheng Ye, Shi-Jim Yen, Julien Dehos, Sergey Zagoruyko, Yu-Jin Lin, Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision (LAMSADE), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), National Taiwan Normal University (NTNU), National Dong Hwa University (NDHU), Université du Littoral Côte d'Opale (ULCO), National Taiwan University [Taiwan] (NTU), Laboratoire d'Informatique Signal et Image de la Côte d'Opale (LISIC), Facebook AI Research [Paris] (FAIR), and Facebook
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, Theoretical computer science, Computer science, Pooling, Computational Mechanics, Machine Learning (stat.ML), 0102 computer and information sciences, 02 engineering and technology, 01 natural sciences, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Machine Learning (cs.LG), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Statistics - Machine Learning, 0202 electrical engineering, electronic engineering, information engineering, Computer Science (miscellaneous), [INFO]Computer Science [cs], Architecture, ComputingMilieux_MISCELLANEOUS, Structure (mathematical logic), Rank (computer programming), Computer Graphics and Computer-Aided Design, Zero (linguistics), Human-Computer Interaction, 010201 computation theory & mathematics, 020201 artificial intelligence & image processing
Abstract: International audience; Since DeepMind's AlphaZero, Zero learning quickly became the state-of-the-art method for many board games. It can be improved using a fully convolutional structure (no fully connected layer). Using such an architecture plus global pooling, we can create bots independent of the board size. The training can be made more robust by keeping track of the best checkpoints during the training and by training against them. Using these features, we release Polygames, our framework for Zero learning, with its library of games and its checkpoints. We won against strong humans at the game of Hex in 19x19 , including the human player with the best ELO rank on LittleGolem; we incidentally also won against another Zero implementation, which was weaker than humans: in a discussion on LittleGolem, Hex19 was said to be intractable for zero learning. We also won in Havannah with size 8: win against the strongest player, namely Eobllor, with excellent opening moves. We also won several first places at the TAAI 2019 competitions and had positive results against strong bots in various games.
Published: 2020
Full Text: View/download PDF

18. Evolving Structures in Complex Systems

Author: Josef Sivic, Tomas Mikolov, Hugo Cisneros, Czech Institute of Informatics, Robotics and Cybernetics [Prague] (CIIRC), Czech Technical University in Prague (CTU), Laboratoire d'informatique de l'école normale supérieure (LIENS), Département d'informatique - ENS Paris (DI-ENS), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), Models of visual object recognition and scene understanding (WILLOW), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria), Facebook AI Research [Paris] (FAIR), Facebook, This work was partially supported by ERC grantLEAP No. 336845, CIFAR Learning in Machines&Brains program and the EU Structural and Investment Funds, Operational Programe Research, Development and Education under the project IMPACT (reg. no. CZ.02.1.01/0.0/0.0/15003/0000468), WILLOW, ANR-19-P3IA-0001,PRAIRIE,PaRis Artificial Intelligence Research InstitutE(2019), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Département d'informatique de l'École normale supérieure (DI-ENS), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS Paris), Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Département d'informatique - ENS Paris (DI-ENS), and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS Paris)
Subjects: FOS: Computer and information sciences, Complex systems, Artificial intelligence, Theoretical computer science, Computer science, Computer Science - Artificial Intelligence, Complex system, FOS: Physical sciences, 02 engineering and technology, Cellular Automata, Emergence, 010501 environmental sciences, 01 natural sciences, Structural complexity, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Artificial life, 0202 electrical engineering, electronic engineering, information engineering, [NLIN.NLIN-CG]Nonlinear Sciences [physics]/Cellular Automata and Lattice Gases [nlin.CG], 0105 earth and related environmental sciences, Computational model, Artificial neural network, Cellular Automata and Lattice Gases (nlin.CG), Complexity, Self-organisation, Grid, Cellular automaton, Artificial Intelligence (cs.AI), Metric (mathematics), 020201 artificial intelligence & image processing, Nonlinear Sciences - Cellular Automata and Lattice Gases
Abstract: In this paper we propose an approach for measuring growth of complexity of emerging patterns in complex systems such as cellular automata. We discuss several ways how a metric for measuring the complexity growth can be defined. This includes approaches based on compression algorithms and artificial neural networks. We believe such a metric can be useful for designing systems that could exhibit open-ended evolution, which itself might be a prerequisite for development of general artificial intelligence. We conduct experiments on 1D and 2D grid worlds and demonstrate that using the proposed metric we can automatically construct computational models with emerging properties similar to those found in the Conway's Game of Life, as well as many other emergent phenomena. Interestingly, some of the patterns we observe resemble forms of artificial life. Our metric of structural complexity growth can be applied to a wide range of complex systems, as it is not limited to cellular automata., IEEE Symposium Series on Computational Intelligence 2019 (IEEE SSCI 2019)
Published: 2019
Full Text: View/download PDF

19. Using Local Knowledge Graph Construction to Scale Seq2Seq Models to Multi-Document Inputs

Author: Chloé Braud, Angela Fan, Antoine Bordes, Claire Gardent, Facebook AI Research [Paris] (FAIR), Facebook, Natural Language Processing : representations, inference and semantics (SYNALP), Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), and Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)
Subjects: FOS: Computer and information sciences, Information retrieval, Computer Science - Computation and Language, Computer science, business.industry, 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, Automatic summarization, Graph, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Ranking, Knowledge base, Knowledge graph, 0202 electrical engineering, electronic engineering, information engineering, Question answering, Graph (abstract data type), 020201 artificial intelligence & image processing, [INFO]Computer Science [cs], business, Computation and Language (cs.CL), Generative grammar, 0105 earth and related environmental sciences
Abstract: International audience; Query-based open-domain NLP tasks require information synthesis from long and diverse web results. Current approaches extrac-tively select portions of web text as input to Sequence-to-Sequence models using methods such as TF-IDF ranking. We propose constructing a local graph structured knowledge base for each query, which compresses the web search information and reduces redundancy. We show that by linearizing the graph into a structured input sequence, models can encode the graph representations within a standard Sequence-to-Sequence setting. For two generative tasks with very long text input , long-form question answering and multi-document summarization, feeding graph representations as input can achieve better performance than using retrieved text portions.
Published: 2019
Full Text: View/download PDF

20. Learning Dynamic Author Representations with Temporal Language Models

Author: Sylvain Lamprier, Ludovic Denoyer, Edouard Delasalles, Machine Learning and Information Access (MLIA), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Facebook AI Research [Paris] (FAIR), and Facebook
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, Computer science, Machine Learning (stat.ML), 02 engineering and technology, Latent variable, 010501 environmental sciences, computer.software_genre, 01 natural sciences, Machine Learning (cs.LG), Text mining, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Statistics - Machine Learning, 0202 electrical engineering, electronic engineering, information engineering, Leverage (statistics), 0105 earth and related environmental sciences, Computer Science - Computation and Language, business.industry, Statistical model, 020201 artificial intelligence & image processing, Language model, Artificial intelligence, business, Computation and Language (cs.CL), computer, Natural language processing
Abstract: Language models are at the heart of numerous works, notably in the text mining and information retrieval communities. These statistical models aim at extracting word distributions, from simple unigram models to recurrent approaches with latent variables that capture subtle dependencies in texts. However, those models are learned from word sequences only, and authors' identities, as well as publication dates, are seldom considered. We propose a neural model, based on recurrent language modeling, which aims at capturing language diffusion tendencies in author communities through time. By conditioning language models with author and temporal vector states, we are able to leverage the latent dependencies between the text contexts. This allows us to beat several temporal and non-temporal language baselines on two real-world corpora, and to learn meaningful author representations that vary through time., International Conference on Data Mining, ICDM 2019
Published: 2019
Full Text: View/download PDF

21. Unsupervised Pre-Training of Image Features on Non-Curated Data

Author: Julien Mairal, Mathilde Caron, Armand Joulin, Piotr Bojanowski, Facebook AI Research [Paris] (FAIR), Facebook, Apprentissage de modèles à partir de données massives (Thoth ), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019]), Department of Statistics [Berkeley], University of California [Berkeley], University of California-University of California, University of California [Berkeley] (UC Berkeley), and University of California (UC)-University of California (UC)
Subjects: FOS: Computer and information sciences, Computer science, Computer Vision and Pattern Recognition (cs.CV), Computer Science - Computer Vision and Pattern Recognition, 02 engineering and technology, 010501 environmental sciences, Machine learning, computer.software_genre, 01 natural sciences, Convolutional neural network, Task (project management), Set (abstract data type), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], 0202 electrical engineering, electronic engineering, information engineering, Code (cryptography), [INFO]Computer Science [cs], Cluster analysis, 0105 earth and related environmental sciences, business.industry, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Feature (computer vision), Unsupervised learning, 020201 artificial intelligence & image processing, Artificial intelligence, business, computer, Feature learning
Abstract: Pre-training general-purpose visual features with convolutional neural networks without relying on annotations is a challenging and important task. Most recent efforts in unsupervised feature learning have focused on either small or highly curated datasets like ImageNet, whereas using uncurated raw datasets was found to decrease the feature quality when evaluated on a transfer task. Our goal is to bridge the performance gap between unsupervised methods trained on curated data, which are costly to obtain, and massive raw datasets that are easily available. To that effect, we propose a new unsupervised approach which leverages self-supervision and clustering to capture complementary statistics from large-scale data. We validate our approach on 96 million images from YFCC100M, achieving state-of-the-art results among unsupervised methods on standard benchmarks, which confirms the potential of unsupervised learning when only uncurated data are available. We also show that pre-training a supervised VGG-16 with our method achieves 74.9% top-1 classification accuracy on the validation set of ImageNet, which is an improvement of +0.8% over the same network trained from scratch. Our code is available at https://github.com/facebookresearch/DeeperCluster., Comment: Accepted at ICCV 2019 (Oral)
Published: 2019
Full Text: View/download PDF

22. Learning to detect dysarthria from raw speech

Author: Neil Zeghidour, Juliette Millet, Laboratoire de Linguistique Formelle (LLF UMR7110), Centre National de la Recherche Scientifique (CNRS)-Université Paris Diderot - Paris 7 (UPD7), Apprentissage machine et développement cognitif (CoML), Laboratoire de sciences cognitives et psycholinguistique (LSCP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS Paris)-École normale supérieure - Paris (ENS Paris)-Centre National de la Recherche Scientifique (CNRS)-École des hautes études en sciences sociales (EHESS)-Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS Paris)-École normale supérieure - Paris (ENS Paris)-Centre National de la Recherche Scientifique (CNRS)-École des hautes études en sciences sociales (EHESS)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Facebook AI Research [Paris] (FAIR), Facebook, Université Paris Diderot - Paris 7 (UPD7)-Centre National de la Recherche Scientifique (CNRS), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Département d'Etudes Cognitives - ENS Paris (DEC), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de sciences cognitives et psycholinguistique (LSCP), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)
Subjects: FOS: Computer and information sciences, Normalization (statistics), Computer Science - Computation and Language, Artificial neural network, Computer science, Speech recognition, Feature extraction, 020206 networking & telecommunications, Data compression ratio, 68T10, 02 engineering and technology, Filter bank, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Raw audio format, Dysarthria, 0202 electrical engineering, electronic engineering, information engineering, medicine, Task analysis, 020201 artificial intelligence & image processing, medicine.symptom, Computation and Language (cs.CL), Classifier (UML)
Abstract: Speech classifiers of paralinguistic traits traditionally learn from diverse hand-crafted low-level features, by selecting the relevant information for the task at hand. We explore an alternative to this selection, by learning jointly the classifier, and the feature extraction. Recent work on speech recognition has shown improved performance over speech features by learning from the waveform. We extend this approach to paralinguistic classification and propose a neural network that can learn a filterbank, a normalization factor and a compression power from the raw speech, jointly with the rest of the architecture. We apply this model to dysarthria detection from sentence-level audio recordings. Starting from a strong attention-based baseline on which mel-filterbanks outperform standard low-level descriptors, we show that learning the filters or the normalization and compression improves over fixed features by 10% absolute accuracy. We also observe a gain over OpenSmile features by learning jointly the feature extraction, the normalization, and the compression factor with the architecture. This constitutes a first attempt at learning jointly all these operations from raw audio for a speech classification task., 5 pages, 3 figures, submitted to ICASSP
Published: 2019

23. Stochastic Adaptive Neural Architecture Search for Keyword Spotting

Author: Tom Veniat, Ludovic Denoyer, Olivier Schwander, Machine Learning and Information Access (MLIA), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Facebook AI Research [Paris] (FAIR), Facebook, and ANR-16-CE23-0016,PAMELA,Apprentissage automatique décentralisé et personnalisé sous contraintes(2016)
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, Computer science, Machine Learning (stat.ML), 02 engineering and technology, 010501 environmental sciences, Machine learning, computer.software_genre, 01 natural sciences, Machine Learning (cs.LG), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Deep Learning, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, Audio and Speech Processing (eess.AS), Statistics - Machine Learning, FOS: Electrical engineering, electronic engineering, information engineering, 0202 electrical engineering, electronic engineering, information engineering, Budgeted Learning, 0105 earth and related environmental sciences, Network architecture, Artificial neural network, business.industry, Deep learning, Neural Architecture Search, Process (computing), Keyword Spotting, Energy consumption, Task (computing), Keyword spotting, 020201 artificial intelligence & image processing, Artificial intelligence, business, computer, Energy (signal processing), Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract: International audience; The problem of keyword spotting i.e. identifying keywords in a real-time audio stream is mainly solved by applying a neural network over successive sliding windows. Due to the difficulty of the task, baseline models are usually large, resulting in a high computational cost and energy consumption level. We propose a new method called SANAS (Stochastic Adaptive Neural Architecture Search) which is able to adapt the architecture of the neural network on-the-fly at inference time such that small architectures will be used when the stream is easy to process (silence, low noise, ...) and bigger networks will be used when the task becomes more difficult. We show that this adaptive model can be learned end-to-end by optimizing a trade-off between the prediction performance and the average computational cost per unit of time. Experiments on the Speech Commands dataset [1] show that this approach leads to a high recognition level while being much faster (and/or energy saving) than classical approaches where the network architecture is static.
Published: 2019
Full Text: View/download PDF

24. Reference-less Quality Estimation of Text Simplification Systems

Author: Pierre-Emmanuel Mazaré, Samuel Humeau, Éric Villemonte de la Clergerie, Antoine Bordes, Benoît Sagot, Louis Martin, Facebook AI Research [Paris] (FAIR), Facebook, Automatic Language Modelling and ANAlysis & Computational Humanities (ALMAnaCH), Inria de Paris, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Machine translation, Computer science, Text simplification, media_common.quotation_subject, 02 engineering and technology, computer.software_genre, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], Task (project management), 03 medical and health sciences, 0302 clinical medicine, 0202 electrical engineering, electronic engineering, information engineering, Quality (business), Simplicity, BLEU, media_common, Computer Science - Computation and Language, business.industry, Reference data, 030221 ophthalmology & optometry, 020201 artificial intelligence & image processing, Grammaticality, Artificial intelligence, business, Computation and Language (cs.CL), computer, Natural language processing
Abstract: International audience; The evaluation of text simplification (TS) systems remains an open challenge. As the task has common points with machine translation (MT), TS is often evaluated using MT metrics such as BLEU. However, such metrics require high quality reference data, which is rarely available for TS. TS has the advantage over MT of being a monolingual task, which allows for direct comparisons to be made between the simplified text and its original version. In this paper, we compare multiple approaches to reference-less quality estimation of sentence-level text simplification systems, based on the dataset used for the QATS 2016 shared task. We distinguish three different dimensions: gram-maticality, meaning preservation and simplicity. We show that n-gram-based MT metrics such as BLEU and METEOR correlate the most with human judgment of grammaticality and meaning preservation, whereas simplicity is best evaluated by basic length-based metrics.
Published: 2019

25. Libri-Light: A Benchmark for ASR with Limited or No Supervision

Author: Armand Joulin, Weiyi Zheng, Morgane Riviere, Ronan Collobert, Tatiana Likhomanenko, Emmanuel Dupoux, Eugene Kharitonov, Pierre-Emmanuel Mazaré, Julien Karadayi, Qiantong Xu, Abdelrahman Mohamed, Jacob Kahn, Vitaliy Liptchinsky, Gabriel Synnaeve, Christian Fuegen, Facebook AI Research [Paris] (FAIR), Facebook, Apprentissage machine et développement cognitif (CoML), Laboratoire de sciences cognitives et psycholinguistique (LSCP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Département d'Etudes Cognitives - ENS Paris (DEC), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), CIFAR program in Learning in Machines & Brains CIFAR LMB program, ANR-19-P3IA-0001,PRAIRIE,PaRis Artificial Intelligence Research InstitutE(2019), ANR-17-EURE-0017,FrontCog,Frontières en cognition(2017), ANR-10-IDEX-0001,PSL,Paris Sciences et Lettres(2010), Facebook AI Research [New York] (FAIR), Distributed Systems Group [Vienne], Technical University of Vienna [Vienna] (TU WIEN), IDIAP Research Institute, Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de sciences cognitives et psycholinguistique (LSCP), École normale supérieure - Paris (ENS-PSL), and Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL)
Subjects: FOS: Computer and information sciences, Sound (cs.SD), Computer science, Speech recognition, Unsupervised and semi-supervised learning, 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, Computer Science - Sound, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], Resource (project management), Audio and Speech Processing (eess.AS), Zero- and low resource ASR, 0202 electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering, 0105 earth and related environmental sciences, Voice activity detection, Computer Science - Computation and Language, Distant supervision, Zero (linguistics), Test (assessment), [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], Benchmark (computing), 020201 artificial intelligence & image processing, Computation and Language (cs.CL), Dataset, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract: International audience; We introduce a new collection of spoken English audio suitable for training speech recognition systems under limited or no supervision. It is derived from open-source audio books from the LibriVox project. It contains over 60K hours of audio , which is, to our knowledge, the largest freely-available corpus of speech. The audio has been segmented using voice activity detection and is tagged with SNR, speaker ID and genre descriptions. Additionally, we provide baseline systems and evaluation metrics working under three settings: (1) the zero resource/unsupervised setting (ABX), (2) the semi-supervised setting (PER, CER) and (3) the distant supervision setting (WER). Settings (2) and (3) use limited textual resources (10 minutes to 10 hours) aligned with the speech. Setting (3) uses large amounts of unaligned text. They are evaluated on the standard LibriSpeech dev and test sets for comparison with the supervised state-of-the-art. Index Terms-unsupervised and semi-supervised learning , distant supervision, dataset, zero-and low resource ASR.
Published: 2019
Full Text: View/download PDF

26. Lead2Gold: Towards exploiting the full potential of noisy transcriptions for speech recognition

Author: Emmanuel Vincent, Matthijs Douze, Armelle Brun, Adrien Dufraux, Awni Hannun, Facebook AI Research [Paris] (FAIR), Facebook, Speech Modeling for Facilitating Oral-Based Communication (MULTISPEECH), Inria Nancy - Grand Est, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Department of Natural Language Processing & Knowledge Discovery (LORIA - NLPKD), Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Laboratoire Lorrain de Recherche en Informatique et ses Applications (LORIA), Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS), Facebook AI Research [New York] (FAIR), Knowledge Information and Web Intelligence (KIWI), Department of Complex Systems, Artificial Intelligence & Robotics (LORIA - AIS), and Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lorraine (UL)-Centre National de la Recherche Scientifique (CNRS)
Subjects: FOS: Computer and information sciences, Computer Science - Machine Learning, Computer science, Computer Science - Artificial Intelligence, Speech recognition, label noise, 02 engineering and technology, noise model, [INFO.INFO-NE]Computer Science [cs]/Neural and Evolutionary Computing [cs.NE], Whole systems, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Machine Learning (cs.LG), 030507 speech-language pathology & audiology, 03 medical and health sciences, Transcription (linguistics), ASR, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), 0202 electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering, beam search, Training set, Computer Science - Computation and Language, Artificial neural network, I.2.6, I.2.7, Artificial Intelligence (cs.AI), [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], Beam search, 020201 artificial intelligence & image processing, 0305 other medical science, Computation and Language (cs.CL), Electrical Engineering and Systems Science - Audio and Speech Processing, weakly supervised learning
Abstract: The transcriptions used to train an Automatic Speech Recognition (ASR) system may contain errors. Usually, either a quality control stage discards transcriptions with too many errors, or the noisy transcriptions are used as is. We introduce Lead2Gold, a method to train an ASR system that exploits the full potential of noisy transcriptions. Based on a noise model of transcription errors, Lead2Gold searches for better transcriptions of the training data with a beam search that takes this noise model into account. The beam search is differentiable and does not require a forced alignment step, thus the whole system is trained end-to-end. Lead2Gold can be viewed as a new loss function that can be used on top of any sequence-to-sequence deep neural network. We conduct proof-of-concept experiments on noisy transcriptions generated from letter corruptions with different noise levels. We show that Lead2Gold obtains a better ASR accuracy than a competitive baseline which does not account for the (artificially-introduced) transcription noise., Comment: 8 pages, 4 tables, Accepted for publication in ASRU 2019
Published: 2019
Full Text: View/download PDF

27. EASSE: Easier Automatic Sentence Simplification Evaluation

Author: Fernando Alva-Manchego, Louis Martin, Lucia Specia, Carolina Scarton, University of Sheffield [Sheffield], Facebook AI Research [Paris] (FAIR), Facebook, Automatic Language Modelling and ANAlysis & Computational Humanities (ALMAnaCH), Inria de Paris, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Computer Science - Computation and Language, Computer science, 02 engineering and technology, computer.software_genre, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], 03 medical and health sciences, 0302 clinical medicine, 030221 ophthalmology & optometry, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Data mining, computer, Computation and Language (cs.CL), Sentence
Abstract: We introduce EASSE, a Python package aiming to facilitate and standardise automatic evaluation and comparison of Sentence Simplification (SS) systems. EASSE provides a single access point to a broad range of evaluation resources: standard automatic metrics for assessing SS outputs (e.g. SARI), word-level accuracy scores for certain simplification transformations, reference-independent quality estimation features (e.g. compression ratio), and standard test data for SS evaluation (e.g. TurkCorpus). Finally, EASSE generates easy-to-visualise reports on the various metrics and features above and on how a particular SS output fares against reference simplifications. Through experiments, we show that these functionalities allow for better comparison and understanding of the performance of SS systems., Comment: EMNLP-IJCNLP 2019 Demo (Camera-ready Version)
Published: 2019
Full Text: View/download PDF

28. Joint Future Semantic and Instance Segmentation Prediction

Author: Pauline Luc, Jakob Verbeek, Camille Couprie, Facebook AI Research [Paris] (FAIR), Facebook, Apprentissage de modèles à partir de données massives (Thoth ), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019]), ANR-16-CE23-0006,Deep_in_France,Réseaux de neurones profonds pour l'apprentissage(2016), and ANR-11-LABX-0025,PERSYVAL-lab,Systemes et Algorithmes Pervasifs au confluent des mondes physique et numérique(2011)
Subjects: Watershed, Computer science, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Optical flow, 02 engineering and technology, 010501 environmental sciences, Color space, Machine learning, computer.software_genre, 01 natural sciences, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Image (mathematics), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Component (UML), 0202 electrical engineering, electronic engineering, information engineering, Segmentation, Representation (mathematics), 0105 earth and related environmental sciences, business.industry, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Key (cryptography), Graph (abstract data type), 020201 artificial intelligence & image processing, Artificial intelligence, business, computer
Abstract: International audience; The ability to predict what will happen next from observing the past is a key component of intelligence. Methods that forecast future frames were recently introduced towards better machine intelligence. However, predicting directly in the image color space seems an overly complex task, and predicting higher level representations using semantic or instance segmentation approaches were shown to be more accurate. In this work, we introduce a novel prediction approach that encodes instance and semantic segmentation information in a single representation based on distance maps. Our graph-based modeling of the instance segmentation prediction problem allows us to obtain temporal tracks of the objects as an optimal solution to a watershed algorithm. Our experimental results on the Cityscapes dataset present state-of-the-art semantic segmentation predictions, and instance segmentation results outperforming a strong baseline based on optical flow.
Published: 2019
Full Text: View/download PDF

29. What Limits Our Capacity to Process Nested Long-Range Dependencies in Sentence Comprehension?

Author: Stanislas Dehaene, Jean-Rémi King, Yair Lakretz, Chaire Psychologie cognitive expérimentale, Collège de France (CdF (institution)), Laboratoire des systèmes perceptifs (LSP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS), Université Paris sciences et lettres (PSL), Centre National de la Recherche Scientifique (CNRS), Facebook AI Research [Paris] (FAIR), Facebook, Neuroimagerie cognitive - Psychologie cognitive expérimentale (UNICOG-U992), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université Paris-Saclay, Collège de France - Chaire Psychologie cognitive expérimentale, École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL), and ANR-17-EURE-0017,FrontCog,Frontières en cognition(2017)
Subjects: Opinion, Computer science, media_common.quotation_subject, General Physics and Astronomy, lcsh:Astrophysics, 02 engineering and technology, computer.software_genre, Psycholinguistics, Sentence processing, 03 medical and health sciences, 0302 clinical medicine, lcsh:QB460-466, 0202 electrical engineering, electronic engineering, information engineering, lcsh:Science, ComputingMilieux_MISCELLANEOUS, media_common, Structure (mathematical logic), Parsing, Grammar, double center-embeddings, business.industry, sentence processing, [SCCO.LING]Cognitive science/Linguistics, lcsh:QC1-999, Comprehension, language model, lcsh:Q, 020201 artificial intelligence & image processing, Artificial intelligence, Language model, business, long-range dependencies, artificial neural networks, computer, lcsh:Physics, 030217 neurology & neurosurgery, Natural language processing, Sentence
Abstract: International audience; Sentence comprehension requires inferring, from a sequence of words, the structure of syntactic relationships that bind these words into a semantic representation. Our limited ability to build some specific syntactic structures, such as nested center-embedded clauses (e.g., “The dog that the cat that the mouse bit chased ran away”), suggests a striking capacity limitation of sentence processing, and thus offers a window to understand how the human brain processes sentences. Here, we review the main hypotheses proposed in psycholinguistics to explain such capacity limitation. We then introduce an alternative approach, derived from our recent work on artificial neural networks optimized for language modeling, and predict that capacity limitation derives from the emergence of sparse and feature-specific syntactic units. Unlike psycholinguistic theories, our neural network-based framework provides precise capacity-limit predictions without making any a priori assumptions about the form of the grammar or parser. Finally, we discuss how our framework may clarify the mechanistic underpinning of language processing and its limitations in the human brain.
Published: 2020
Full Text: View/download PDF

30. Dense Pose Transfer

Author: Riza Alp Guler, Natalia Neverova, Iasonas Kokkinos, Facebook AI Research [Paris] (FAIR), Facebook, Centre de vision numérique (CVN), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec, Organ Modeling through Extraction, Representation and Understanding of Medical Image Content (GALEN-POST), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), and University College of London [London] (UCL)
Subjects: FOS: Computer and information sciences, Surface (mathematics), Pixel, business.industry, Computer science, Computer Vision and Pattern Recognition (cs.CV), Coordinate system, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Computer Science - Computer Vision and Pattern Recognition, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, Pipeline (software), Image (mathematics), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Computer vision, Artificial intelligence, Image warping, business, Pose, 0105 earth and related environmental sciences
Abstract: In this work we integrate ideas from surface-based modeling with neural synthesis: we propose a combination of surface-based pose estimation and deep generative models that allows us to perform accurate pose transfer, i.e. synthesize a new image of a person based on a single image of that person and the image of a pose donor. We use a dense pose estimation system that maps pixels from both images to a common surface-based coordinate system, allowing the two images to be brought in correspondence with each other. We inpaint and refine the source image intensities in the surface coordinate system, prior to warping them onto the target pose. These predictions are fused with those of a convolutional predictive module through a neural synthesis module allowing for training the whole pipeline jointly end-to-end, optimizing a combination of adversarial and perceptual losses. We show that dense pose estimation is a substantially more powerful conditioning input than landmark-, or mask-based alternatives, and report systematic improvements over state of the art generators on DeepFashion and MVC datasets., ECCV 2018
Published: 2018

31. End-to-End Speech Recognition from the Raw Waveform

Author: Ronan Collobert, Neil Zeghidour, Gabriel Synnaeve, Nicolas Usunier, Emmanuel Dupoux, Université Paris-Saclay, Apprentissage machine et développement cognitif (CoML), Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de sciences cognitives et psycholinguistique (LSCP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Département d'Etudes Cognitives - ENS Paris (DEC), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), Facebook AI Research [Paris] (FAIR), Facebook, Laboratoire d'Informatique de Grenoble (LIG ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019]), IDIAP Research Institute, Laboratoire de sciences cognitives et psycholinguistique (LSCP), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Normalization (statistics), Sound (cs.SD), Vocabulary, Computer science, Speech recognition, Pipeline (computing), media_common.quotation_subject, Normalization (image processing), Initialization, Word error rate, 02 engineering and technology, Signal, [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL], Computer Science - Sound, [SCCO]Cognitive science, 030507 speech-language pathology & audiology, 03 medical and health sciences, Audio and Speech Processing (eess.AS), 020204 information systems, FOS: Electrical engineering, electronic engineering, information engineering, 0202 electrical engineering, electronic engineering, information engineering, Waveform, Index Terms: speech recognition, media_common, Computer Science - Computation and Language, scattering, Filter (signal processing), end-to-end, [SCCO.LING]Cognitive science/Linguistics, waveform, deep, [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], gammatones, 0305 other medical science, Computation and Language (cs.CL), Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract: State-of-the-art speech recognition systems rely on fixed, hand-crafted features such as mel-filterbanks to preprocess the waveform before the training pipeline. In this paper, we study end-to-end systems trained directly from the raw waveform, building on two alternatives for trainable replacements of mel-filterbanks that use a convolutional architecture. The first one is inspired by gammatone filterbanks (Hoshen et al., 2015; Sainath et al, 2015), and the second one by the scattering transform (Zeghidour et al., 2017). We propose two modifications to these architectures and systematically compare them to mel-filterbanks, on the Wall Street Journal dataset. The first modification is the addition of an instance normalization layer, which greatly improves on the gammatone-based trainable filterbanks and speeds up the training of the scattering-based filterbanks. The second one relates to the low-pass filter used in these approaches. These modifications consistently improve performances for both approaches, and remove the need for a careful initialization in scattering-based trainable filterbanks. In particular, we show a consistent improvement in word error rate of the trainable filterbanks relatively to comparable mel-filterbanks. It is the first time end-to-end models trained from the raw signal significantly outperform mel-filterbanks on a large vocabulary task under clean recording conditions., Accepted for presentation at Interspeech 2018
Published: 2018
Full Text: View/download PDF

32. Deforming Autoencoders: Unsupervised Disentangling of Shape and Appearance

Author: Dimitris Samaras, Mihir Sahasrabudhe, Iasonas Kokkinos, Nikos Paragios, Riza Alp Guler, Zhixin Shu, Department of Computer Science [Stonybrook - NY], Stony Brook University [SUNY] (SBU), State University of New York (SUNY)-State University of New York (SUNY), Centre de vision numérique (CVN), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec, Organ Modeling through Extraction, Representation and Understanding of Medical Image Content (GALEN-POST), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Facebook AI Research [Paris] (FAIR), and Facebook
Subjects: FOS: Computer and information sciences, Computer science, Computer Vision and Pattern Recognition (cs.CV), Coordinate system, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Computer Science - Computer Vision and Pattern Recognition, 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, Image (mathematics), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, 0202 electrical engineering, electronic engineering, information engineering, [INFO.INFO-IM]Computer Science [cs]/Medical Imaging, 0105 earth and related environmental sciences, ComputingMethodologies_COMPUTERGRAPHICS, business.industry, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Pattern recognition, Expression (mathematics), Generative model, Face (geometry), [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV], Computer Science::Computer Vision and Pattern Recognition, 020201 artificial intelligence & image processing, Artificial intelligence, business, Interpolation
Abstract: In this work we introduce Deforming Autoencoders, a generative model for images that disentangles shape from appearance in an unsupervised manner. As in the deformable template paradigm, shape is represented as a deformation between a canonical coordinate system (`template') and an observed image, while appearance is modeled in `canonical', template, coordinates, thus discarding variability due to deformations. We introduce novel techniques that allow this approach to be deployed in the setting of autoencoders and show that this method can be used for unsupervised group-wise image alignment. We show experiments with expression morphing in humans, hands, and digits, face manipulation, such as shape and appearance interpolation, as well as unsupervised landmark localization. A more powerful form of unsupervised disentangling becomes possible in template coordinates, allowing us to successfully decompose face images into shading and albedo, and further manipulate face images., 17 pages including references, plus 12 pages appendix. Video available at : https://youtu.be/Oi7pyxKkF1g Code will be made available soon
Published: 2018

33. DensePose: Dense Human Pose Estimation in the Wild

Author: Iasonas Kokkinos, Riza Alp Guler, Natalia Neverova, Centre de vision numérique (CVN), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec, Organ Modeling through Extraction, Representation and Understanding of Medical Image Content (GALEN-POST), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Institut National de Recherche en Informatique et en Automatique (Inria), Facebook AI Research [Paris] (FAIR), Facebook, and University College of London [London] (UCL)
Subjects: FOS: Computer and information sciences, Ground truth, Computer science, business.industry, Computer Vision and Pattern Recognition (cs.CV), Computer Science - Computer Vision and Pattern Recognition, Inpainting, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], 020207 software engineering, 02 engineering and technology, Frame rate, Pipeline (software), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], 0202 electrical engineering, electronic engineering, information engineering, Code (cryptography), 020201 artificial intelligence & image processing, Computer vision, Artificial intelligence, Scale (map), business, Pose
Abstract: International audience; In this work, we establish dense correspondences between an RGB image and a surface-based representation of the human body, a task we refer to as dense human pose estimation. We gather dense correspondences for 50K persons appearing in the COCO dataset by introducing an efficient annotation pipeline. We then use our dataset to train CNN-based systems that deliver dense correspondence ‘in the wild’, namely in the presence of background, occlusions and scale variations. We improve our training set’s effectiveness by training an inpainting network that can fill in missing ground truth values and report improvements with respect to the best results that would be achievable in the past. We experiment with fully-convolutional networks and region-based models and observe a superiority of the latter. We further improve accuracy through cascading, obtaininga system that delivers highly-accurate results at multiple frames per second on a single GPU. Supplementary materials, data, code, and videos are provided on the project page http://densepose.org.
Published: 2018
Full Text: View/download PDF

34. Link and code: Fast indexing with graphs and compact regression codes

Author: Alexandre Sablayrolles, Hervé Jégou, Matthijs Douze, Facebook AI Research [Paris] (FAIR), Facebook, Apprentissage de modèles à partir de données massives (Thoth ), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019]), Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Inria Grenoble - Rhône-Alpes, and Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Theoretical computer science, Computer science, Computer Vision and Pattern Recognition (cs.CV), Nearest neighbor search, Computer Science - Computer Vision and Pattern Recognition, 02 engineering and technology, Computer Science - Information Retrieval, Computer Science - Databases, 020204 information systems, Graph traversal, Computer Science - Data Structures and Algorithms, 0202 electrical engineering, electronic engineering, information engineering, Data Structures and Algorithms (cs.DS), [INFO]Computer Science [cs], business.industry, Quantization (signal processing), Search engine indexing, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Databases (cs.DB), Graph, [INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR], 020201 artificial intelligence & image processing, Artificial intelligence, business, Information Retrieval (cs.IR)
Abstract: International audience; Similarity search approaches based on graph walks have recently attained outstanding speed-accuracy trade-offs, taking aside the memory requirements. In this paper, we revisit these approaches by considering, additionally, the memory constraint required to index billions of images on a single server. This leads us to propose a method based both on graph traversal and compact representations. We encode the indexed vectors using quantization and exploit the graph structure to refine the similarity estimation. In essence, our method takes the best of these two worlds: the search strategy is based on nested graphs, thereby providing high precision with a relatively small set of comparisons. At the same time it offers a significant memory compression. As a result, our approach outperforms the state of the art on operating points considering 64–128 bytes per vector, as demonstrated by our results on two billion-scale public benchmarks.
Published: 2018
Full Text: View/download PDF

35. Predicting Future Instance Segmentation by Forecasting Convolutional Features

Author: Yann LeCun, Pauline Luc, Jakob Verbeek, Camille Couprie, Facebook AI Research [Paris] (FAIR), Facebook, Courant Institute of Mathematical Sciences [New York] (CIMS), New York University [New York] (NYU), NYU System (NYU)-NYU System (NYU), Apprentissage de modèles à partir de données massives (Thoth ), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019]), ANR-16-CE23-0006,Deep_in_France,Réseaux de neurones profonds pour l'apprentissage(2016), ANR-11-LABX-0025,PERSYVAL-lab,Systemes et Algorithmes Pervasifs au confluent des mondes physique et numérique(2011), Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Inria Grenoble - Rhône-Alpes, and Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Computer science, Computer Vision and Pattern Recognition (cs.CV), Optical flow, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Computer Science - Computer Vision and Pattern Recognition, 02 engineering and technology, 010501 environmental sciences, Machine learning, computer.software_genre, 01 natural sciences, Convolutional neural network, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Video Prediction, 0202 electrical engineering, electronic engineering, information engineering, Segmentation, Instance Segmentation, 0105 earth and related environmental sciences, business.industry, Deep learning, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], RGB color model, 020201 artificial intelligence & image processing, Convolutional neural networks, Artificial intelligence, business, computer
Abstract: International audience; Anticipating future events is an important prerequisite towards intelligent behavior. Video forecasting has been studied as a proxy task towards this goal. Recent work has shown that to predict semantic segmentation of future frames, forecasting at the semantic level is more effective than forecasting RGB frames and then segmenting these. In this paper we consider the more challenging problem of future instance segmentation, which additionally segments out individual objects. To deal with a varying number of output labels per image, we develop a predictive model in the space of fixed-sized convolutional features of the Mask R-CNN instance segmentation model. We apply the "detection head" of Mask R-CNN on the predicted features to produce the instance segmentation of future frames. Experiments show that this approach significantly improves over strong baselines based on optical flow and repurposed instance segmentation architectures.
Published: 2018
Full Text: View/download PDF

36. Object Level Visual Reasoning in Videos

Author: Fabien Baradel, Natalia Neverova, Christian Wolf, Julien Mille, Greg Mori, Extraction de Caractéristiques et Identification (imagine), Laboratoire d'InfoRmatique en Image et Systèmes d'information (LIRIS), Institut National des Sciences Appliquées de Lyon (INSA Lyon), Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-École Centrale de Lyon (ECL), Université de Lyon-Université Lumière - Lyon 2 (UL2)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Université Lumière - Lyon 2 (UL2), Facebook AI Research [Paris] (FAIR), Facebook, Robots coopératifs et adaptés à la présence humaine en environnements dynamiques (CHROMA), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-CITI Centre of Innovation in Telecommunications and Integration of services (CITI), Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon, Laboratoire d'Informatique Fondamentale et Appliquée de Tours (LIFAT), Université de Tours (UT)-Institut National des Sciences Appliquées - Centre Val de Loire (INSA CVL), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), Department of Computer Science (Simon Fraser University), Simon Fraser University (SFU.ca), Wolf, Christian, Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA), Centre National de la Recherche Scientifique (CNRS)-Université de Tours-Institut National des Sciences Appliquées - Centre Val de Loire (INSA CVL), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA), Université Lumière - Lyon 2 (UL2)-École Centrale de Lyon (ECL), Université de Lyon-Université de Lyon-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Université Lumière - Lyon 2 (UL2)-École Centrale de Lyon (ECL), and Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)
Subjects: FOS: Computer and information sciences, Computer science, Computer Vision and Pattern Recognition (cs.CV), Video understanding, Computer Science - Computer Vision and Pattern Recognition, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Context (language use), 02 engineering and technology, Visual reasoning, 010501 environmental sciences, 16. Peace & justice, Object (computer science), 01 natural sciences, Motion (physics), Object detection, Activity recognition, [INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Human–computer interaction, 0202 electrical engineering, electronic engineering, information engineering, Key (cryptography), 020201 artificial intelligence & image processing, Human-object interaction, 0105 earth and related environmental sciences
Abstract: Human activity recognition is typically addressed by detecting key concepts like global and local motion, features related to object classes present in the scene, as well as features related to the global context. The next open challenges in activity recognition require a level of understanding that pushes beyond this and call for models with capabilities for fine distinction and detailed comprehension of interactions between actors and objects in a scene. We propose a model capable of learning to reason about semantically meaningful spatiotemporal interactions in videos. The key to our approach is a choice of performing this reasoning at the object level through the integration of state of the art object detection networks. This allows the model to learn detailed spatial interactions that exist at a semantic, object-interaction relevant level. We evaluate our method on three standard datasets (Twenty-BN Something-Something, VLOG and EPIC Kitchens) and achieve state of the art results on all of them. Finally, we show visualizations of the interactions learned by the model, which illustrate object classes and their interactions corresponding to different activity classes., Comment: Accepted at ECCV 2018 - long version (16 pages + ref)
Published: 2018
Full Text: View/download PDF

37. Dense and Low-Rank Gaussian CRFs Using Deep Embeddings

Author: Iasonas Kokkinos, Nicolas Usunier, Siddhartha Chandra, Organ Modeling through Extraction, Representation and Understanding of Medical Image Content (GALEN), Ecole Centrale Paris-Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Facebook AI Research [Paris] (FAIR), Facebook, Inria Saclay - Ile de France, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Ecole Centrale Paris
Subjects: Conditional random field, Computational complexity theory, Computer science, business.industry, Gaussian, Inference, 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], symbols.namesake, Conjugate gradient method, 0202 electrical engineering, electronic engineering, information engineering, symbols, 020201 artificial intelligence & image processing, Pairwise comparison, Artificial intelligence, Structured prediction, business, Algorithm, 0105 earth and related environmental sciences, Sparse matrix
Abstract: International audience; In this work we introduce a structured prediction model that endows the Deep Gaussian Conditional Random Field (G-CRF) with a densely connected graph structure. We keep memory and computational complexity under control by expressing the pairwise interactions as inner products of low-dimensional, learnable embeddings. The G-CRF system matrix is therefore low-rank, allowing us to solve the resulting system in a few milliseconds on the GPU by using conjugate gradient. As in G-CRF, inference is exact, the unary and pairwise terms are jointly trained end-to-end by using analytic expressions for the gradients, while we also develop even faster, Potts-type variants of our embeddings. We show that the learned embeddings capture pixel-to-pixel affinities in a task-specific manner, while our approach achieves state of the art results on three challenging benchmarks, namely semantic segmentation, human part segmentation, and saliency estimation. Our implementation is fully GPU based, built on top of the Caffe library, and is available at https://github.com/ siddharthachandra/gcrf-v2.0.
Published: 2017

38. Predicting Deeper into the Future of Semantic Segmentation

Author: Jakob Verbeek, Yann LeCun, Natalia Neverova, Camille Couprie, Pauline Luc, Facebook AI Research [Paris] (FAIR), Facebook, Apprentissage de modèles à partir de données massives (Thoth ), Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria), New York University [New York] (NYU), NYU System (NYU), ANR-11-LABX-0025,PERSYVAL-lab,Systemes et Algorithmes Pervasifs au confluent des mondes physique et numérique(2011), ANR-16-CE23-0006,Deep_in_France,Réseaux de neurones profonds pour l'apprentissage(2016), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK ), and Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])
Subjects: FOS: Computer and information sciences, Computer science, Computer Vision and Pattern Recognition (cs.CV), Optical flow, Computer Science - Computer Vision and Pattern Recognition, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, 02 engineering and technology, 010501 environmental sciences, Machine learning, computer.software_genre, 01 natural sciences, Convolutional neural network, Machine Learning (cs.LG), 0202 electrical engineering, electronic engineering, information engineering, Segmentation, Image warping, 0105 earth and related environmental sciences, Pixel, business.industry, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Pattern recognition, Image segmentation, Object detection, Computer Science - Learning, RGB color model, 020201 artificial intelligence & image processing, Artificial intelligence, business, computer
Abstract: The ability to predict and therefore to anticipate the future is an important attribute of intelligence. It is also of utmost importance in real-time systems, e.g. in robotics or autonomous driving, which depend on visual scene understanding for decision making. While prediction of the raw RGB pixel values in future video frames has been studied in previous work, here we introduce the novel task of predicting semantic segmentations of future frames. Given a sequence of video frames, our goal is to predict segmentation maps of not yet observed video frames that lie up to a second or further in the future. We develop an autoregressive convolutional neural network that learns to iteratively generate multiple frames. Our results on the Cityscapes dataset show that directly predicting future segmentations is substantially better than predicting and then segmenting future RGB frames. Prediction results up to half a second in the future are visually convincing and are much more accurate than those of a baseline based on warping semantic segmentations using optical flow., Comment: Accepted to ICCV 2017. Supplementary material available on the authors' webpages
Published: 2017
Full Text: View/download PDF

39. Convolutional Patch Representations for Image Retrieval: an Unsupervised Approach

Author: Matthijs Douze, Florent Perronnin, Julien Mairal, Mattis Paulin, Zaid Harchaoui, Cordelia Schmid, Apprentissage de modèles à partir de données massives (Thoth ), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019]), Facebook AI Research [Paris] (FAIR), Facebook, New York University [New York] (NYU), NYU System (NYU), ANR-14-CE23-0003,MACARON,Apprentissage statistique à grande échelle et applications(2014), Laboratoire Jean Kuntzmann (LJK ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes [2016-2019] (UGA [2016-2019])-Inria Grenoble - Rhône-Alpes, and Institut National de Recherche en Informatique et en Automatique (Inria)
Subjects: FOS: Computer and information sciences, Computer science, Computer Vision and Pattern Recognition (cs.CV), Computer Science - Computer Vision and Pattern Recognition, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Scale-invariant feature transform, Model parameters, Low-level image description, 02 engineering and technology, 010501 environmental sciences, Machine learning, computer.software_genre, 01 natural sciences, Convolutional neural network, Artificial Intelligence, Instance-level retrieval, 0202 electrical engineering, electronic engineering, information engineering, Visual Word, Image retrieval, 0105 earth and related environmental sciences, Contextual image classification, business.industry, Convolutional Neural Networks, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Pattern recognition, Kernel (image processing), Computer Science::Computer Vision and Pattern Recognition, 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Artificial intelligence, business, computer, Software
Abstract: International audience; Convolutional neural networks (CNNs) have recently received a lot of attention due to their ability to model local stationary structures in natural images in a multi-scale fashion, when learning all model parameters with supervision. While excellent performance was achieved for image classification when large amounts of labeled visual data are available, their success for un-supervised tasks such as image retrieval has been moderate so far. Our paper focuses on this latter setting and explores several methods for learning patch descriptors without supervision with application to matching and instance-level retrieval. To that effect, we propose a new family of convolutional descriptors for patch representation , based on the recently introduced convolutional kernel networks. We show that our descriptor, named Patch-CKN, performs better than SIFT as well as other convolutional networks learned by artificially introducing supervision and is significantly faster to train. To demonstrate its effectiveness, we perform an extensive evaluation on standard benchmarks for patch and image retrieval where we obtain state-of-the-art results. We also introduce a new dataset called RomePatches, which allows to simultaneously study descriptor performance for patch and image retrieval.
Published: 2017
Full Text: View/download PDF

40. Composing Relationships with Translations

Author: Alberto Garcia-Duran, Nicolas Usunier, Antoine Bordes, Heuristique et Diagnostic des Systèmes Complexes [Compiègne] (Heudiasyc), Université de Technologie de Compiègne (UTC)-Centre National de la Recherche Scientifique (CNRS), Facebook AI Research [Paris] (FAIR), Facebook, and CNRS, Heudiasyc
Subjects: Theoretical computer science, business.industry, Computer science, 020204 information systems, 0202 electrical engineering, electronic engineering, information engineering, Embedding, Graph (abstract data type), 020201 artificial intelligence & image processing, [INFO]Computer Science [cs], 02 engineering and technology, Artificial intelligence, business, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract: International audience; Performing link prediction in Knowledge Bases (KBs) with embedding-based models , like with the model TransE (Bordes et al., 2013) which represents relationships as translations in the embedding space, have shown promising results in recent years. Most of these works focused on modeling single relationships and hence do not take full advantage of the graph structure of KBs. In this paper, we propose an extension of TransE that learns to explicitly model composition of relationships via the addition of their corresponding translation vectors. We show empirically that this allows to improve performance for predicting single relationships as well as compositions of pairs of them.
Published: 2015
Full Text: View/download PDF

41. Speech Resynthesis from Discrete Disentangled Self-Supervised Representations

Author: Emmanuel Dupoux, Jade Copet, Wei-Ning Hsu, Kushal Lakhotia, Eugene Kharitonov, Yossi Adi, Abdelrahman Mohamed, Adam Polyak, Facebook AI Research [Paris] (FAIR), Facebook, Laboratoire de sciences cognitives et psycholinguistique (LSCP), Département d'Etudes Cognitives - ENS Paris (DEC), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS), Apprentissage machine et développement cognitif (CoML), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Département d'Etudes Cognitives - ENS Paris (DEC), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École des hautes études en sciences sociales (EHESS)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), ANR-19-P3IA-0001,PRAIRIE,PaRis Artificial Intelligence Research InstitutE(2019), ANR-17-EURE-0017,FrontCog,Frontières en cognition(2017), ANR-10-IDEX-0001,PSL,Paris Sciences et Lettres(2010), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS Paris), Inria de Paris, and Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de sciences cognitives et psycholinguistique (LSCP)
Subjects: FOS: Computer and information sciences, Self-supervised learning, Sound (cs.SD), Computer Science - Machine Learning, Computer science, media_common.quotation_subject, Speech recognition, Speech coding, Speech generation, 02 engineering and technology, Intelligibility (communication), Identity (music), Computer Science - Sound, Machine Learning (cs.LG), Task (project management), Speech codec, 030507 speech-language pathology & audiology, 03 medical and health sciences, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, 0202 electrical engineering, electronic engineering, information engineering, Quality (business), Representation (mathematics), Speech resynthesis, media_common, Speech quality, 020206 networking & telecommunications, [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], 0305 other medical science, Feature learning, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract: We propose using self-supervised discrete representations for the task of speech resynthesis. To generate disentangled representation, we separately extract low-bitrate representations for speech content, prosodic information, and speaker identity. This allows to synthesize speech in a controllable manner. We analyze various state-of-the-art, self-supervised representation learning methods and shed light on the advantages of each method while considering reconstruction quality and disentanglement properties. Specifically, we evaluate the F0 reconstruction, speaker identification performance (for both resynthesis and voice conversion), recordings' intelligibility, and overall quality using subjective human evaluation. Lastly, we demonstrate how these representations can be used for an ultra-lightweight speech codec. Using the obtained representations, we can get to a rate of 365 bits per second while providing better speech quality than the baseline methods. Audio samples can be found under the following link: speechbot.github.io/resynthesis., In Proceedings of Interspeech 2021

Catalog

Books, media, physical & digital resources

See catalog results

Searchworks

Select search scope, currently: Articles Catalog books, media & more in Jio Institute collections Articles journal articles & other e-resources

Search

Search Constraints

Refine your results

Search Limiters

Topic

Publication Year Range

Language

Journal

Database

Publisher

41 results on '"Facebook AI Research [Paris] (FAIR)"'

Search Results

Catalog

Select search scope, currently: Articles

Catalog

books, media & more in Jio Institute collections

Articles

journal articles & other e-resources